Perfecting fiscal decentralization to increase economic growth in Vietnam

This paper identifies negative impacts of budget expenditure decentralization on Vietnam’s economic growth and positive impacts of central government expenditure, private investments and trade openness on the economic growth as well. Additionally, no relationship between inflation rate along with changes in labor force and economic growth is found.

Perfecting Fiscal Decentralization to Increase

Economic Growth in Vietnam

SỬ ĐÌNH THÀNH

Associate Professor, Doctor of Philosophy, University of Economics - HCMC

Email: dinhthanh@ueh.edu.vn MAI ĐÌNH LÂM

Master of Arts, Ho Chi Minh National Academy of Politics and Public Administration

Email: maidinhlam2006@yahoo.com

ABSTRACT

The government intervention in the economy is practical and widely acknowledged The central government expenditure, like budget incomes, is to regulate the national economy According to Keynes (1936), the government should aim at demand-side stimuli to facilitate consumption and production This paper identifies negative impacts of budget expenditure decentralization on Vietnam’s economic growth; and positive impacts of central government expenditure, private investments and trade openness on the economic growth as well Additionally, no relationship between inflation rate along with changes in labor force and economic growth is found

Keywords: economic growth, budget expenditure, private investment, trade openness, labor force, inflation

1 INTRODUCTION

In order to promote economic growth, budget expenditures by authorities of any levels must be thoroughly weighed up Fiscal decentralization and empowerment of lower authorities are part of public section reform that aims at enhancing competitiveness of lower authorities in provision of public services and getting out of the clumsiness of the economy (Bahl & Linn, 1992; Bird, Ebel & Wallich, 1993) In the world there have been plenty of researches on impacts of budget expenditure decentralization on economic growth In Vietnam, some examples are those by Hoàng Thị Chinh Thon et al (2010) and Phạm Thế Anh (2008) Yet, their papers just dealt with a specific locality but not the whole national economy Hence, the present paper is

to investigate such impact on Vietnam’s national economy as a whole

2 LITERATURE REVIEW

There have been numerous researches on impacts of budget expenditure decentralization on economic growth such as that of Mankiw, Romer and Weil (1992) which failed to manifest the role of government expenditure in economic growth Kormedi and Meguire (1985) and Barro (1991) did employ data of many different economies and utilized multiple regressions to explain different growth rates in surveyed countries in the long term; and variables were opted in accordance with growth theories and speculations But these two papers produced different results Kormendi and Meguire (1985) contended that government expenditure had no impact

on the economic growth while Barro (1991) proved vice versa

Davoodi, Swaroop and Zou (1996), using data from 43 countries and in more than

20 years of researching, indicated that enhancement of investment expenditure had negative impacts on the economic growth whereas increases in recurrent expenditure had positive impacts Ghosh and Gregoriou (2008) employed the generalized method

of moments (GMM) to analyze the data collated from 15 developing countries in a 28-year period and obtained the same results As their empirical research indicated, recurrent expenditure but not investment expenditure was significant to economic growth

Nguyễn Phi Lân (2008) analyzed the data of 34 provinces and cities of Vietnam in the period 2000-2005 by the parametric approach (based on the random production function) and the non-parametric approach (based on DEA) He posited that the ineffectiveness of public expenditure existed in annual public expenditure and

investment Phạm Thế Anh(2008) employed data of 61 provinces and cities of Vietnam in the period 2001-2005; and analyzed investment expenditure and recurrent expenditure in five different industries The research indicated more positive impacts

of investment expenditure compared with recurrent expenditure in some industries and vice versa in other fields

Hoàng Thị Chinh Thon et al (2010) studied how public expenditure at province and district levels influenced the local economic growth by a regression model Using data collated from 31 localities in Vietnam, she showed that expenditure at district level should be increased and that at province level should be decreased so as to stimulate the local economic growth

3 RESEARCH MODEL

In this paper, the author employed the data collated in the period 1990-2011 and the neoclassical production function with expansion of endogenous variables Indeed, if the technical factor (A) is ignored, the comprehensive production function can be simplified as below:

Y = f(K, L) (1)

Where, Y denotes the production yield; K is the private investment; and L represents the labor force

Regarding capital formation and stimulation of aggregate demand, Ram (1986), with the time series data of 115 countries, employed the extended production function

associated with a variable of government total expenditure (G) to reach the conclusion that central government expenditure (G) was just one of factors affecting the economic

growth Moreover, Yingyi Qian and Meredith Woo-Cumings, when studying the government and public sector reform in South Korea, as quoted by Yusuf and Stiglitz (2002), emphasized the role of public expenditure in promoting private investment and realizing development targets Like previous researches, this one included the variable

of government expenditure (G) as an independent input factor related to formation of

capital needed for economic growth

Y = f(K, L, G) (2)

Concerning fiscal decentralization, G is divided into central government expenditure (TW) and local authorities expenditure (DF); and thus: G = TW + DF

To evaluate impacts of fiscal decentralization on economic growth, local budget

expenditures can be divided into investment expenditure (DF I) and recurrent

expenditure (DF C) We have:

DF = DF I + DF C

And thus: Y = f(K, L, TW, DF I , DF C )

The variable of inflation rate (lp) can be added to the model to evaluate its impacts

on the economic growth, and the trade openness (xnk) to evaluate the international

integration of Vietnam’s economy Then the gross production function of six macroeconomic variables can be rewritten as follows:

Y = f(K, L,TW, DF I , DF C , lp, xnk) (3)

Taking the derivative the function (3) of Y (excluding the inflation rate - lp), we

have the equation (4) as follows:

lp dlp lp Y L dL L Y Y dxnk

xnk

Y

Y dDF DF Y Y dDF DF Y Y dTW TW Y Y dK K Y

Y

/ ) / ( / ) / ( / )

/

(

/ ) / ( / ) / ( / ) / ( / ) /

(

/











































(4)

where /Y  K,  /Y  lp,  /Y  TW,  /Y  DF Iand  /Y  DF C are respectively the marginal factor of capital, inflation, central government expenditure, investment expenditure, and recurrent expenditure in comparison with GDP  /Y  xnk and

L

Y 

 / are respectively the marginal factor of trade openness and labor force

4 RELATIONSHIP BETWEEN FISCAL DECENTRALIZATION AND ECONOMIC GROWTH

Pursuant to the applicable National Budget Law, local authorities mainly assume responsibility for provision of public services within the locality When assigning expenditure responsibilities, it is expected that local budget expenditures might generate a higher demand side in the local economy and thereby promoting the local and national economic growth

Figure 1: The Relationship between the Ratio of Local Budget Expenditure to

GDP and the GDP Growth Rate (as %)

Source: GSO and Ministry of Finance, 1990-2011

It is possible to preliminarily evaluate the relationship between budget expenditure delegation and economic growth as follows:

- For the period 1990 – 1996: delegation of budget expenditure to local authorities reached a high of 19% of GDP in 1990, which then fell in 1992; yet Vietnam’s economic growth was quite high, over 8% In the period 1994 – 1996, GDP was rising steadily while the ratio of local budget expenditure to GDP went down In short, in this period, the relationship between budget expenditure delegation and economic growth was negative

- For the period 1997 – 2003: After the 2006 National Budget Law, the ratio of budget expenditure delegated soared swiftly The economic growth reached 8.8% in

1997 and then gradually fell in the following years This accounted for inefficiency in budget expenditures, especially recurrent ones In this period, the Asian economic recession also profoundly influenced Vietnamese economy The economic growth pace only recovered in 2002 and 2003 to 7% and 7.3% respectively

- For the period 2004 – 2011: The ratio of local budget expenditure to GDP grew unsteadily The GDP in this period was quite high, 7.8% in 2004, 8.44% in 2005, 8.23% in 2006 and 8.46% in 2007 Yet in 2008, the growth rate was merely 6.31% due

to the recession of the world economy; and then was falling in the following years This indicated that although local authorities were provided with more expenditure responsibilities, budget incomes were not properly assigned, causing the central government to constantly make up for local budgets It is a weakness of budget decentralization in Vietnam In this period, the relationship between assignment of

0,00

10,00

20,00

30,00

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

gdp Chi ĐP/GDP Ratio of local budget expenditure to GDP

expenditure responsibilities and economic growth was negative, which will be clarified

in following sections of the paper

The local budget expenditures are also divided into local investment expenditures and local recurrent expenditures Figure 2 indicates the relationship between the ratio

of local investment expenditure to GDP and economic growth rate

Figure 2: The Relationship between the Ratio of Local Investment Expenditure to

GDP and the Economic Growth Rate (as %)

Source: GSO and Ministry of Finance, 1990-2011

As Figure 2 shows, the local investment expenditure in the period 1990 – 2000 was financed by local budget incomes and balanced revenues, or by targeted transfers from central government determined by the GDP growth rate This indicates that public expenditures by local authorities are still a fundamental factor to promote the economic growth in the years 1990-2000 In the next decade (2001 – 2011), however, the trend

of local investment expenditure and GDP growth rate was opposite, which might be explained by the ineffectiveness of local investment expenditures and high ICOR; and thus it could not stimulate the economic development

It is questioned whether or not it is necessary to change the scale of local expenditures with a view to improving the economic growth In fact, Vietnam’s economic growth, as from 1997, cannot be merely explained by the fiscal decentralization It is generally admitted that Vietnam, after the 1997 financial crisis, has adopted many dynamic policies to stimulate the economic growth such as: encouraging private business, attracting foreign investments, promoting foreign trade and liberating the financial system, etc

0,00

5,00

10,00

15,00

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

gdp Chi ĐTĐP/GDP Ratio of local investment expenditure to GDP

Figure 3: The Relationship Between the Ratio of Local Recurrent Expenditure to

GDP and the Economic Growth Rate (as %)

Source: GSO and Ministry of Finance, 1990-2011

Additionally, private investment, trade openness and inflation are also able to promote or hinder the economic growth

5 RESEARCH MODEL AND RESULTS

a Research Model:

Before estimation, the equation (4) is rewritten as:

lp dlp lp Y L dL L Y Y dxnk

xnk

Y

Y dDF DF Y Y dDF DF Y Y dTW TW Y Y dK K Y

Y

/ ) / ( / ) / ( / )

/

(

/ ) / ( / ) / ( / ) / ( / ) /

(

/









































where, dY/Y is the annual GDP growth rate; dK/Y, I/Y or PI is the ratio of private investment to GDP; dL/L or PGR is the labor force fluctuation ratio; dlp/lp or inf is the inflation rate; dTW/Y or CG is the ratio of central government expenditure to GDP; dDF/Y or LG is the ratio of local budget expenditure to GDP; dDF I /Y or LG I is the

local investment expenditure to GDP; dDF C /Y or LG C is the ratio of local recurrent

expenditure to GDP; dxnk/Y or TOP is the ratio of total export and import turnover to

GDP which is used to measure the economic openness

With:

1

6

After adjustment, equation (5) is rewritten as follows:

t t

C t

I t t

t o

GDP 1 2 3 4 5 6 7inf

0,00

10,00

20,00

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 201

gdp Chi TXĐP/GDP Ratio of local recurrent expenditure to GDP

The equation (5) indicates that the economic growth depends on variables CG, LG I ,

LG C , PI, PGR, inf and TOP

In order to test the research mode, the statistical equation (6) below is employed

t t t

t

C t

I t t

t o

GDP  1 2 3 4 5 6 7inf  (6)

In the model, the yearly data are mainly secondary numerical data Specifically, economic growth rates are collated from GSO; local budget expenditure and income from the Ministry of Finance and the official website of Vietnam’s government (www.chinhphu.vn); private investments, economic openness and inflation rate from the Ministry of Industry and Commerce and GSO, website of Vietnam’s General Department of Customs, Asian Development Bank, and author’s computation Such data are summarized in Table 1

Table 1: Mean Values Stat

b Test Results:

- Stationarity testing:

The Augmented Dickey – Fuller (ADF) is employed to test whether or not the time series data set is stationary Results are in Table 2

Table 2: Testing Stationarity of Variables

N.B.: ***, **, * respectively denote the statistical significance level of 10%, 5% and 1%

As Table 2 indicates, the time series data set of GDP is stationary with the

significance level of 10% Due to the fact that the number of observations is too small, the significance level of 10% is deemed as an acceptable standard in the present

research Also, LG, LG C , and CG are stationary with the significance level of 5%; while PGR at 1% and LG I at 10% Due to the fact that TOP and PI are not stationary at the acceptable significance level, the authors find their simple difference (dTOP and dPI) to test their stationarity; and the results show that dTOP and dPI are respectively

stationary at the significance of 1% and 5%

- Testing results:

With

t t t

t

C t

I t t

t o

GDP  1 2 3 4 5 6 7inf  The OLS method is employed to evaluate the relationship between annual GDP

growth rate and independent variables LG and CG Others including PI, PGR, inf and TOP play as control variables The evaluation results are presented in Table 3

Table 3: Estimation Results

Dependent variable: GDP Variable Coefficient

Standard error t-stat ρ-value

Adjusted R-squared 0.735331 S.D dependent var 1.343045 S.E of regression 0.690943 Akaike info criterion 2.367697 Sum squared resid 6.206221 Schwarz criterion 2.716203 Log likelihood -16.67697 Hannan-Quinn criter 2.435729

As Table 3 indicates, at the significance level of 10%, most variables are statistically significant, excluding the inflation rate and population growth rate; and this is consistent with the theoretical background The inflation rate has a bidirectional relationship with economic growth rate, that is, it can accompany with a high economic growth rate due to the loose fiscal and monetary policies, and it can also hinder the economic growth rate when the fiscal and monetary policies are tightened The population growth rate might not influence the economic growth rate of a developing country like Vietnam because the full employment is not secured and that the unemployment rate, both official and quasi-official one, is very high